ISO 639 - New item approved - N'Ko
Caoimhin O Donnaile
caoimhin at smo.uhi.ac.uk
Fri Jun 9 23:42:12 CEST 2006
> Regardless of which which Unicode character is ultimately selected for
> the third glyph in the subject language, the ASCII (and presumably
> English) name N'Ko should be registered too.
I don't think the "and presumably English" can be assumed these days.
It might have been true a few years ago, but we are now, or very
soon will be, in the Unicode era and I don't think it is any longer
true that "English" can be written in ASCII, nor even in ISO 8859-1,
because of all the apostrophes.
Nevertheless, "ASCIIfied English", if we can call it that, might still
be useful, and one possible policy would be for the registry to store as
Descriptions, for every language:
- the ASCIIfied English name
- the native name (only if different from the above?) in full Unicode glory
Although this might be useful for searching, though, it might be
confusing for applications using the registry data - unless the
various Descriptions could easily be distinguished from one another
(e.g. by order, if this can be assumed to be defined?)
> Why is it an problem to have 6 alternative ASCII forms of Micmac, if
> there really isn't a definitive name for the language in English?
I suppose again, that it would be good for searching, but possibly
bad for applications using the registry data if they couldn't easily
decide which of the various names it would be preferable to use.
Another possible policy would be to say that the registry can't do
everything, and that all that is needed in the Description(s) is just
enough to attempt to uniquely distinguish the language from others.
For everything else, users would be referred to external resources.
I think that applications dealing with possibly thousands of languages
would need to give the user some kind of "language chooser" applet based
on geography and/or language family, rather than relying on purely
(still using a pre-Unicode era mailer)
More information about the Ietf-languages